Using Social Media to Help Understand Patient-Reported Health Outcomes of Post-COVID-19 Condition: Natural Language Processing Approach.

BACKGROUND: While scientific knowledge of post-COVID-19 condition (PCC) is growing, there remains significant uncertainty in the definition of the disease, its expected clinical course, and its impact on daily functioning. Social media platforms can generate valuable insights into patient-reported health outcomes as the content is produced at high resolution by patients and caregivers, representing experiences that may be unavailable to most clinicians. OBJECTIVE: In this study, we aimed to determine the validity and effectiveness of advanced natural language processing approaches built to derive insight into PCC-related patient-reported health outcomes from social media platforms Twitter and Reddit. We extracted PCC-related terms, including symptoms and conditions, and measured their occurrence frequency. We compared the outputs with human annotations and clinical outcomes and tracked symptom and condition term occurrences over time and locations to explore the pipeline's potential as a surveillance tool. METHODS: We used bidirectional encoder representations from transformers (BERT) models to extract and normalize PCC symptom and condition terms from English posts on Twitter and Reddit. We compared 2 named entity recognition models and implemented a 2-step normalization task to map extracted terms to unique concepts in standardized terminology. The normalization steps were done using a semantic search approach with BERT biencoders. We evaluated the effectiveness of BERT models in extracting the terms using a human-annotated corpus and a proximity-based score. We also compared the validity and reliability of the extracted and normalized terms to a web-based survey with more than 3000 participants from several countries. RESULTS: UmlsBERT-Clinical had the highest accuracy in predicting entities closest to those extracted by human annotators. Based on our findings, the top 3 most commonly occurring groups of PCC symptom and condition terms were systemic (such as fatigue), neuropsychiatric (such as anxiety and brain fog), and respiratory (such as shortness of breath). In addition, we also found novel symptom and condition terms that had not been categorized in previous studies, such as infection and pain. Regarding the co-occurring symptoms, the pair of fatigue and headaches was among the most co-occurring term pairs across both platforms. Based on the temporal analysis, the neuropsychiatric terms were the most prevalent, followed by the systemic category, on both social media platforms. Our spatial analysis concluded that 42% (10,938/26,247) of the analyzed terms included location information, with the majority coming from the United States, United Kingdom, and Canada. CONCLUSIONS: The outcome of our social media-derived pipeline is comparable with the results of peer-reviewed articles relevant to PCC symptoms. Overall, this study provides unique insights into patient-reported health outcomes of PCC and valuable information about the patient's journey that can help health care providers anticipate future needs. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): RR2-10.1101/2022.12.14.22283419.

Figure Correction: Using Social Media to Help Understand Patient-Reported Health Outcomes of Post-COVID-19 Condition: Natural Language Processing Approach.

[This corrects the article DOI: 10.2196/45767.].

Machine learning hypothesis-generation for patient stratification and target discovery in rare disease: our experience with Open Science in ALS.

Introduction: Advances in machine learning (ML) methodologies, combined with multidisciplinary collaborations across biological and physical sciences, has the potential to propel drug discovery and development. Open Science fosters this collaboration by releasing datasets and methods into the public space; however, further education and widespread acceptance and adoption of Open Science approaches are necessary to tackle the plethora of known disease states. Motivation: In addition to providing much needed insights into potential therapeutic protein targets, we also aim to demonstrate that small patient datasets have the potential to provide insights that usually require many samples (>5,000). There are many such datasets available and novel advancements in ML can provide valuable insights from these patient datasets. Problem statement: Using a public dataset made available by patient advocacy group AnswerALS and a multidisciplinary Open Science approach with a systems biology augmented ML technology, we aim to validate previously reported drug targets in ALS and provide novel insights about ALS subpopulations and potential drug targets using a unique combination of ML methods and graph theory. Methodology: We use NetraAI to generate hypotheses about specific patient subpopulations, which were then refined and validated through a combination of ML techniques, systems biology methods, and expert input. Results: We extracted 8 target classes, each comprising of several genes that shed light into ALS pathophysiology and represent new avenues for treatment. These target classes are broadly categorized as inflammation, epigenetic, heat shock, neuromuscular junction, autophagy, apoptosis, axonal transport, and excitotoxicity. These findings are not mutually exclusive, and instead represent a systematic view of ALS pathophysiology. Based on these findings, we suggest that simultaneous targeting of ALS has the potential to mitigate ALS progression, with the plausibility of maintaining and sustaining an improved quality of life (QoL) for ALS patients. Even further, we identified subpopulations based on disease onset. Conclusion: In the spirit of Open Science, this work aims to bridge the knowledge gap in ALS pathophysiology to aid in diagnostic, prognostic, and therapeutic strategies and pave the way for the development of personalized treatments tailored to the individual's needs.

Cone-beam computed tomography for on-line image guidance of lung stereotactic radiotherapy: localization, verification, and intrafraction tumor position.

PURPOSE: Cone-beam computed tomography (CBCT) in-room imaging allows accurate inter- and intrafraction target localization in stereotactic body radiotherapy of lung tumors. METHODS AND MATERIALS: Image-guided stereotactic body radiotherapy was performed in 28 patients (89 fractions) with medically inoperable Stage T1-T2 non-small-cell lung carcinoma. The targets from the CBCT and planning data set (helical or four-dimensional CT) were matched on-line to determine the couch shift required for target localization. Matching based on the bony anatomy was also performed retrospectively. Verification of target localization was done using either megavoltage portal imaging or CBCT imaging; repeat CBCT imaging was used to assess the intrafraction tumor position. RESULTS: The mean three-dimensional tumor motion for patients with upper lesions (n = 21) and mid-lobe or lower lobe lesions (n = 7) was 4.2 and 6.7 mm, respectively. The mean difference between the target and bony anatomy matching using CBCT was 6.8 mm (SD, 4.9, maximum, 30.3); the difference exceeded 13.9 mm in 10% of the treatment fractions. The mean residual error after target localization using CBCT imaging was 1.9 mm (SD, 1.1, maximum, 4.4). The mean intrafraction tumor deviation was significantly greater (5.3 mm vs. 2.2 mm) when the interval between localization and repeat CBCT imaging (n = 8) exceeded 34 min. CONCLUSION: In-room volumetric imaging, such as CBCT, is essential for target localization accuracy in lung stereotactic body radiotherapy. Imaging that relies on bony anatomy as a surrogate of the target may provide erroneous results in both localization and verification.

Geometric performance and efficiency of an optical tracking system for daily pre-treatment positioning in pelvic radiotherapy patients.

The purpose of this study was to characterize the accuracy of a novel in-house optical tracking system (OTS), and to determine its efficiency for daily pre-treatment positioning of pelvic radiotherapy patients compared to conventional optical distance indicator (ODI) methodology. The OTS is comprised of a passive infrared stereoscopic camera, and custom control software for use in assisting radiotherapy patient setup. Initially, the system was calibrated and tested for stability inside a radiation therapy treatment room. Subsequently, under an ethics approved protocol, the clinical efficiency of the OTS was compared to conventional ODI setup methodology through 17 pelvic radiotherapy patients. Differences between orthogonal source-to-skin distance (SSD) readings and overall set-up time resultant from both systems were compared. The precision of the OTS was 0.01 ± 0.01 mm, 0.02 ± 0.02 mm, and -0.01 ± 0.06 mm in the left/right (L/R), anterior/posterior (A/P), and cranial/caudal (C/C) directions, respectively. Discrepancies measured between the linac radiographic center in the treatment room and the calibrated origin of the camera (OTS) by two independent observers was submillimeter. Analysis of 146 fractions from 17 patients showed a high correlation between the SSD readings of the OTS and ODI setup methodologies (r = 0.99). The average time for pre-treatment positioning using the OTS couch shift calculation was 2.60 ± 0.69 minutes, and for conventional ODI setup, 3.62 ± 0.82 minutes; the difference of 1.02 minutes was statistically significant (p < 0.001). In conclusion, the OTS is a precise and robust tool for use as an independent check of treatment room patient positioning. The system is indicated as geometrically equivalent to current methods of daily pre-treatment patient positioning with potential for gains in efficiency by decreasing setup times in the treatment room.